In the Mad Laboratory

The problem​ with owning an invisible book is that it’s hard to lay your hands on it when you need it. This I know from experience, having recently searched the bookshelves at length for mine, which I bought several years ago from its creator and author, Elisabeth Tonnard. It cost €0, and I have an invoice as proof of purchase. It was one of a limited edition of a hundred, neither signed nor numbered; as Tonnard’s website states, it’s ‘a product without a single fault, available at the lowest price possible’. Some may scoff, but it turned out to be a shrewd investment, since it is now sold out, and used copies are changing hands on eBay for larger sums. There is also an accompanying set of postcards, ‘Highlights in the History of The Invisible Book’, which includes pictures of the book’s early underwater testing in the Galápagos Islands, its acclaimed 1962 exhibition at the New York Public Library, and the undisclosed facility where the original manuscript has been kept since the 1870s, though ‘some say it is no longer there.’ (It was, ironically, these visible components that I couldn’t find, hiding between the volumes in the bookcase.)

Tonnard’s Invisible Book may be amusing, but it’s more than a joke. We’re increasingly comfortable with the idea of a book in virtual rather than physical form, but what happens when the content disappears too? Inevitably, we’re left looking at the frame around it. In an exhibition at the Royal Library of the Netherlands in The Hague in 2019, visitors were confronted with a caption labelling an apparently empty glass case, but there are many less literal frames that surround books: blurbs, publishing information, reviews, library catalogues, academic articles and all manner of other paratexts that contribute to the literary artefact. You’re reading one right now. The point about the invisible book, in other words, is that it suddenly makes other things visible: a whole cultural apparatus normally taken for granted which Annette Gilbert, in Literature’s Elsewheres, calls the ‘conditions of literature’. Gilbert’s new study places Tonnard’s book in a long line of strange attempts to stretch and warp the parameters of writing. As a field it’s disparate and hard to summarise, easier to define through what it doesn’t do than what it does. These are works that don’t create fictional worlds or try to represent reality. Instead they ‘reflect upon and performatively test the actual, literal conditions of their existence’.

It’s all very meta, like Vito Acconci’s ‘[READ THIS WORD THEN READ THIS WORD]’ (1969), a text that comprises nothing more than a series of step-by-step instructions for reading it. Or Dan Graham’s ‘Poem’, first published in 1966-67 in the art magazine Aspen, whose contents are in effect a comprehensive list of the text’s features, from the number of adjectives and lines to the paper stock on which it is printed and the typeface used. This means that it’s endlessly variable, different every time it’s reproduced (the poet Craig Dworkin even commissioned a version on a rug). Robert Barry’s ‘The Space between Pages’ calls attention to its own material location in a different way. Carried in 1969 by the New York avant-garde periodical 0 TO 9, it was listed on the contents page, but readers searched for it in vain. The magazine explained that ‘the advertised work is not located on a specific page, but between the specified pages.’ The pages in question – 29 and 30 – are the front and back of a single leaf. Tonnard’s is not the first invisible or ‘non-retinal’ work of literature.

Acconci, Graham and Barry are names familiar from the American avant-garde scene of the late 1960s and 1970s, who borrowed ideas from conceptual and site-specific art but transferred them into print. They reasoned that since art circulated largely on paper anyway – in the form of exhibition catalogues, reproductions and reviews – why not cut out the middleman? The page itself could be a ‘field for action’ – a site of experimentation rather than an inert background. It could even be an exhibition space, as with Seth Siegelaub’s Xerox Book, described as a show that ran not ‘for three weeks’, but for ‘175 pages’. This late 1960s milieu understandably looms large in Gilbert’s study, yet this isn’t a survey of a particular moment – or movement – in art. It’s not about literature either, or not exactly. It’s about what happens in the hinterland between the two. The ‘elsewheres’ in the title is not just a reference to the elusive, disappearing nature of some of this work, but to its location in a cultural sense, on the ‘cusp’ or the ‘removes’ of literature as a discipline.

On these outer limits, things are more permissive. Tonnard may now be categorised as an artist, but mostly by default. Initially she tried her luck as a writer, sending Bloodaxe some of her ‘erasure’ poetry, which she wrote by removing words. She got a rejection letter that used the same technique: ‘Thank … send … Eliot’s … sorry.’ In 1912 Gertrude Stein got the same treatment, receiving a letter caricaturing her distinctively stuttering, repetitive poetry: ‘Only one look, only one look is enough. Hardly one copy would sell here. Hardly one. Hardly one.’ The experimentation Gilbert describes is of a different order, not with writing but with the page, the book and the notion of the literary itself. This is confusingly hybrid territory, with no shortage of portmanteau labels – artists’ books, conceptual poetry, bookworks, pageworks – but Gilbert doesn’t get too bogged down in the terminology. Instead she traces a loose lineage of work that crosses customary divides of discipline and period from the 1950s onwards, but which has precursors in Mallarmé’s typographic experiments and even in the book that first defined the idea of the meta-novel, Tristram Shandy.

Despite Gilbert’s occasionally ponderous register – ‘instantiated entities’ and ‘praxeological challenge’ – this stuff is a riot, a mad laboratory, exploring questions you’re not usually allowed to ask. What happens if you turn an ebook back into a book, for instance? The answer isn’t as obvious as it seems. The digitised text of Heart of Darkness, ‘poured’ into a paperback by Stephanie Syjuco, becomes almost illegible, the text so chopped up and erratically distributed that Conrad’s slim novel becomes a 400-page concrete poem, scattered with indecipherable characters and lines of computer code. A book can have more than one cover, but can it have more than one title? Elfriede Jelinek’s wir sind lockvögel baby! (1970) tries this out, sporting a clear plastic pocket with a removable cardboard label bearing the name of the author and the title of the book, which readers can exchange with six alternative title plates. Nanni Balestrini’s novel Tristano (1966) takes the idea a step further, asking if the same book can vary on the inside rather than the outside. No two copies have exactly the same plot. Each one is unique, since it’s ‘written’ using a computer algorithm which rearranges the same passages in any number of possible combinations.

Nick Thurston’s Of the Subcontract. Or Principles of Poetic Right (2013) outsources its authorship to anonymous gig workers. Its hundred poems, all written to order and purchased via Amazon’s Mechanical Turk online platform, are reproduced alongside details of the hourly rate, time spent and total paid for each, ranging from one cent to one dollar. (Even the foreword, attributed to the media theorist McKenzie Wark, was supplied by a ghost writer from Lahore, commissioned through freelancer.com for $75.) The resulting doggerel is sometimes surprisingly frank in its response to the task at hand:

0.76. Work
00:00:21 → $130.29/HR → 1/1

Writing is such a hard job
Much rather words I would rob
Those people on ’Turk
They’ll do all the work
So my brain I do not have to prod.

Thurston’s name on the cover isn’t a conventional form of literary attribution, but part of the provocation, as if to illustrate the Mexican theorist Ulises Carrión’s much cited idea that nowadays authors don’t write texts, they make books. The creative act isn’t producing, but publishing and ‘curating’.

And what if the text you’re publishing has already been published by someone else, and you’re simply reissuing it under your own name, as Sherrie Levine does with Flaubert’s Un coeur simple? (She graciously thanks Flaubert in her acknowledgments.) Is this simply theft, or a transformation into something new? The question was posed by Borges in his story ‘Pierre Menard, Author of the Quixote’ (1939), whose protagonist sets out to write – rather than merely copy – a few chapters of Don Quixote, with his text corresponding to the original word for word. An identical passage – banal and trite when written by Cervantes – becomes ‘astounding’ in its insight when Menard re-authors it some 350 years later. But what if, in another turn of the screw, someone were themselves to write Menard’s opus, appropriating this appropriation? Inevitably, they have, more than once. The artist Elaine Sturtevant has published the work of Pierre Menard under her own name, bringing this fictional book into the real world as Sturtevant: Author of the ‘Quixote’ (2009). It’s a brain twister, as Gilbert points out, since Sturtevant ‘puts herself not only in the place of Cervantes, but also in the place of Menard, whose Quixote project, conceptualised by Borges, she actually implements with this book’. Meanwhile, Aurélie Noury’s publishing imprint, Lorem Ipsum, dedicates itself entirely to the production of ‘imaginary books lodged inside real books’, producing flesh and blood editions not only of Menard’s Quixote but also of six other fictional tomes described by Borges, in collaboration with his friend Adolfo Bioy Casares, in ‘A List and Analysis of the Sundry Books of F.J.C. Loomis’ (1969), a story about a fictional writer that purports to have been written by another fictional writer, Bustos Domecq.

These are the flipside of Tonnard’s Invisible Book – books that have a material existence, but shouldn’t. They present some intractable problems of categorisation: on what side of the fact/fiction divide do their various elements belong? Is the non-existent F.J.C. Loomis the author of his poems, or is Noury, as publisher? The Bibliothèque nationale de France resolves the issue by attributing authorship to Loomis, but describing him as an ‘auteur fictif’. These literary elsewheres often seem specifically intended to cause headaches for librarians and cataloguers, as with Fiona Banner’s imprint Vanity Press, which has registered a host of non-book objects with an ISBN. The items include Banner herself (she has the number tattooed on her back), meaning that the British Library is entitled to ask for a copy of her.

This is a kind of ‘institutional critique’ in which libraries, universities and the publishing industry are the focus of experiment as much as books themselves, with grit thrown into the gears so that their workings are laid bare. But perhaps it is only possible because the institution of literature is looking precarious. Gilbert writes that ‘the age of the bookish text is gradually coming to a close,’ and points to ‘discursive shifts’ that have destabilised its conventions. In contrast to earlier readers, she claims, we now have a ‘generally heightened sensitivity to the form of the text and the book’, attending to typographic irregularities and other oddities. It’s here that her argument falls down. Try telling this to scholars of the early modern period: the book was then a much wilder and weirder thing. With untrustworthy or absent paratexts, irregularity and error as the norm, and the conventions of the printed book inventing themselves before their very eyes, readers in earlier centuries had no option but to be attuned to the quirks of the page. It could be argued that it’s the modern reader, accustomed to the uniformity of the book in the industrial era, who tends to read past any strangenesses unless jolted out of the habit.

Johanna Drucker​ ’s Inventing the Alphabet is all about writing’s material histories. It doesn’t get more material than the story of Yaqoub Caravacca, a Jordanian boy who in 1868 was hired to make a papier-mâché impression of a stone inscription, at which point he was lanced through the leg by angry Bedouins. The stone was the Mesha Stele, a huge basalt slab inscribed with linear Phoenician – an early alphabetic script – in the mid-ninth century BCE. Its ownership was disputed by French, British and German archaeologists, and the local population, who broke it into pieces, which they used to bless their crops. The writing surface has been making its presence felt for millennia, for readers as well as writers: stone, metal, papyrus, wood, parchment, paper. These distinctions are crucial for the evolution of the letters you’re reading now. Curvy lines only became possible with the movements of a pen or stylus, quite unlike angular Roman capitals or the spiky horizontals and notches of the early medieval Ogham script, both of which were carved into stone. The alphabet has always been embedded in tools and technologies, and still undergirds global communication in the form of Unicode, the alphanumeric system on which the internet relies.

Although it began as a single invention, the alphabet has been reconfigured in every era, and given a host of different origin stories. Herodotus, writing in the fifth century BCE, claimed it was a gift from the Phoenician Cadmus to the Greeks. The legend has a grain of truth to it: the latest archaeological discoveries point to a piecemeal emergence in the Middle East some three and a half thousand years ago through a process of cultural exchange, as migrant Semitic-speaking labourers adapted Egyptian hieroglyphs as phonetic symbols, and these were gradually standardised and began to spread northwards across the Mediterranean. Arabic, runes, Greek, Devanagari, Hebrew and Roman letters are all derivations of this original script. But we don’t get to this authoritative account until we’ve waded through thickets of citation and speculation from earlier scholars. Drucker spends much of her time in earlier centuries, among gentleman antiquarians, kabbalists and scribes, following strange intellectual tangents and cul-de-sacs.

The 17th and 18th centuries, which were much preoccupied with the alphabet’s origins, put forward a variety of possible sources: Teutons, Anglo-Saxons, Celts and even the Belgians. But things could get tangled. Richard Verstegan’s A Restitution of Decayed Intelligence in Antiquities concerning the Most Noble and Renowned English Nation (1605) asserted that the Irish tongue was originally German, and was spoken by Adam and Eve, ‘the etymology of whose names he teased out in imaginative fashion to show that they were one person – “even” – meaning “the same as” because “Eve” was Adam’s wife’. This kind of cod etymology and linguistic nationalism evolved into something darker in the 19th century with attempts to establish the alphabet as a uniquely Hellenic and thus European invention. And in the 20th century its history was co-opted by racialist discourse, by a strain of antisemitism that Aryanised and warped its lineage, editing the Hebrew alphabet out of the story.

‘Alphabet studies’, as Drucker terms this field, has always been a magnet for ideological fixations, monomania and crankery. Religious belief, too, has been woven into its fabric. According to the Bible the alphabet had a divine origin, given to Moses in the form of the Ten Commandments. But this only created more questions and paradoxes. How was Moses able to decipher the letters? Did writing exist beforehand? In the early 18th century, Daniel Defoe argued with faultless logic that the answer was no, since all kinds of things would then have been written down that hadn’t been. Venus, for instance, would have been exposed as an ‘everlasting Whore, an insatiate impudent Strumpet’. This merging of biblical story and classical myth as historical evidence shows that Defoe wasn’t overly concerned with such distinctions, but alphabet studies have always accommodated mixtures of the factual, mythical and mystical. The 16th-century vogue for compendiums was fuelled by curiosity about exotic and ancient scripts, but these real if unfamiliar symbols were supplemented by wholly imagined ones. There were so-called ‘Ring letters’, with circles on their tips that had magical properties, and ‘celestial alphabets’ dictated by angels, including one known as ‘Crossing the River’, which appeared in the alchemist Johannes Pantheus’s Voarchadumia (1530), and was reprinted in one compendium after another. These existed mostly as exemplars – graphic images – and couldn’t be used to write with, though they were sometimes used for other purposes. The Enochian alphabet was ‘revealed’ by angels to the magician John Dee in 1583, who soon put it to use in his conjuring activities.

These apocryphal alphabets – unreadable, speculative experiments on the borders of writing – wouldn’t be out of place in Literature’s Elsewheres. There’s much common ground between Drucker and Gilbert, despite their very different subject matter. Both are authorities on the ‘artist’s book’ – Drucker wrote the now standard survey, The Century of Artists’ Books (1995) – and both recognise that what they’re describing isn’t a coherent and discrete field, but a sprawling and complex area of study pieced together from wayward experiments or the discarded ideas of earlier centuries. If Gilbert’s book presents writing from the subversive and playful perspective of art, Drucker sees it through the multiple lenses of alchemy, cryptography, theology, epigraphy (the deciphering of text in stone) and so on. These, too, are ‘elsewheres’ – places where the premodern imagination is unfettered by the boundaries of knowledge, where writing is made strange, displaced, turned inside out.